CN106156794B - Character recognition method and device based on character style recognition - Google Patents
Character recognition method and device based on character style recognition Download PDFInfo
- Publication number
- CN106156794B CN106156794B CN201610509781.6A CN201610509781A CN106156794B CN 106156794 B CN106156794 B CN 106156794B CN 201610509781 A CN201610509781 A CN 201610509781A CN 106156794 B CN106156794 B CN 106156794B
- Authority
- CN
- China
- Prior art keywords
- character
- recognition
- style
- text
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
Abstract
The invention provides a character recognition method and a device based on character style recognition, wherein the character recognition method comprises the following steps: recognizing the character style of an input character image, and outputting character style information associated with the character image; and selecting a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for carrying out character recognition on the character image. According to the character recognition method and device based on character style recognition, the character style is recognized before character recognition is carried out, the character recognition database of the character style is selected for character recognition based on different character styles, and therefore recognition efficiency can be improved, and recognition accuracy can be improved.
Description
Technical Field
The invention relates to the technical field of character recognition, in particular to a character recognition method and device based on character style recognition.
Background
Character recognition is a technique of automatically recognizing characters using a computer. At present, the word recognition based on the neural network achieves a high recognition rate, and the general work flow is as follows: extracting an input picture signal into a tensor form; extracting and deforming the character part in the picture as the input of a character recognition network; and outputting the identification result at the output node after traversing the nodes of the neural network according to the network structure.
However, the existing neural network-based character recognition system has no variability, and the same recognition mode is adopted for the inputs of different character styles, which not only may lead to the reduction of the recognition rate, but also is not beneficial to the improvement of the recognition efficiency. For example, existing text recognition systems may not be distinguishable for different text that has similar appearance due to different fonts; for another example, the shape of the same character is different due to different fonts, and the existing character recognition system may erroneously recognize the same character as a different character.
Therefore, new technical means are required to solve the above problems.
Disclosure of Invention
The present invention has been made in view of the above problems. The invention provides a character recognition method and device based on character style recognition, which dynamically optimizes a character recognition system by using a character recognition method assisted by character style recognition and improves the recognition accuracy, thereby improving the existing single recognition system.
According to an aspect of the present invention, there is provided a character recognition method based on character style recognition, the character recognition method including: recognizing the character style of an input character image, and outputting character style information associated with the character image; and selecting a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for carrying out character recognition on the character image.
In one embodiment of the invention, the text style comprises at least one of: the font of the characters, the language of the characters and the object for presenting the characters.
In one embodiment of the present invention, the recognition of the character style of the input character image is based on a trained neural network.
In one embodiment of the present invention, the text style information includes information of a plurality of selectable text styles similar to the text style in the text image.
In one embodiment of the present invention, the selecting the character recognition database corresponding to the character style information includes selecting the character recognition database corresponding to the information of the character style with the highest similarity of character styles in the character image.
According to another aspect of the present invention, there is provided a character recognition apparatus based on character style recognition, the character recognition apparatus including: the character style recognition module is used for recognizing the character style of the input character image and outputting character style information associated with the character image; and the character recognition module is used for selecting a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for carrying out character recognition on the character image.
In one embodiment of the invention, the text style comprises at least one of: the font of the characters, the language of the characters and the object for presenting the characters.
In one embodiment of the invention, the character style recognition module performs character style recognition on the input character image based on the trained neural network.
In one embodiment of the present invention, the text style information includes information of a plurality of selectable text styles similar to the text style in the text image.
In an embodiment of the present invention, the character recognition module selects a character recognition database corresponding to information of a character style with the highest similarity to the character style in the character image for character recognition of the character image.
According to the character recognition method and device based on character style recognition, the character style is recognized before character recognition is carried out, the character recognition database of the character style is selected for character recognition based on different character styles, and therefore recognition efficiency can be improved, and recognition accuracy can be improved.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in more detail embodiments of the present invention with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings, like reference numbers generally represent like parts or steps.
FIG. 1 is a schematic block diagram of an example electronic device for implementing a text-style-recognition-based text recognition method and apparatus in accordance with embodiments of the present invention;
FIG. 2 is a schematic flow chart of a method of text recognition based on text style recognition in accordance with an embodiment of the present invention;
FIG. 3 is a schematic block diagram of a text recognition apparatus based on text style recognition according to an embodiment of the present invention; and
FIG. 4 is a schematic block diagram of a text recognition system based on text style recognition in accordance with an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of embodiments of the invention and not all embodiments of the invention, with the understanding that the invention is not limited to the example embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the invention described herein without inventive step, shall fall within the scope of protection of the invention.
First, an exemplary electronic device 100 for implementing a text-style-recognition-based text recognition method and apparatus according to an embodiment of the present invention is described with reference to fig. 1.
As shown in FIG. 1, electronic device 100 includes one or more processors 102, one or more memory devices 104, an input device 106, an output device 108, and an image sensor 110, which are interconnected via a bus system 112 and/or other form of connection mechanism (not shown). It should be noted that the components and structure of the electronic device 100 shown in fig. 1 are exemplary only, and not limiting, and the electronic device may have other components and structures as desired.
The processor 102 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.
The storage 104 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. On which one or more computer program instructions may be stored that may be executed by processor 102 to implement client-side functionality (implemented by the processor) and/or other desired functionality in embodiments of the invention described below. Various applications and various data, such as various data used and/or generated by the applications, may also be stored in the computer-readable storage medium.
The input device 106 may be a device used by a user to input instructions and may include one or more of a keyboard, a mouse, a microphone, a touch screen, and the like.
The output device 108 may output various information (e.g., images or sounds) to an external (e.g., user), and may include one or more of a display, a speaker, and the like.
The image sensor 110 may take images (e.g., photographs, videos, etc.) desired by the user and store the taken images in the storage device 104 for use by other components.
Exemplary electronic devices for implementing the method and apparatus for recognizing words based on word style recognition according to embodiments of the present invention may be implemented as, for example, smart phones, tablet computers, and the like.
Next, a character recognition method 200 based on character style recognition according to an embodiment of the present invention will be described with reference to fig. 2.
In step S210, character style recognition is performed on the input character image, and character style information associated with the character image is output.
In one embodiment, the input text image may be a text image captured by the image capturing device, or a text image from another source. The text image may be a text picture, a text video, or the like. The text style of the text image may include, but is not limited to, the font of the text, the language of the text, the object on which the text is presented, various types of artistic words, and the like. The object presenting the text is, for example, paper, stone, wood, or other possible objects.
For characters with different character styles, even the same character can have different appearances, and different characters can have similar appearances due to different character styles. Taking the font as an example, the date in a relatively flat font is difficult to distinguish from the date in a common font. Therefore, the recognition rate can be effectively improved by recognizing the character style before the character recognition.
In one embodiment, the character style of the input character image can be recognized based on the trained neural network. Taking the font as an example, the font recognition can be performed on the input text image based on the trained neural network. Illustratively, training a neural network for font recognition may be, for example: acquiring a large number of character images, labeling character fonts in the character images, extracting font characteristics of characters in the character images, and training a classification model for font recognition on a training set by adopting a deep learning algorithm (such as a random forest algorithm) based on the extracted font characteristics, namely training a neural network for font recognition. The neural network is convenient to train, stable in performance and strong in universality, so that the character style of the input character image is recognized simply, effectively and easily on the basis of the trained neural network. In other embodiments, the input text image may also be subject to text style recognition based on classical Optical Character Recognition (OCR).
Through the recognition of the character style, the character style information associated with the input character image can be output. For example, taking a font as an example, font information associated with an input text image may be output through font recognition of the input text image. Similarly, for example, by performing language identification on the input text image, language information associated with the input text image may be output.
In one embodiment, the recognition of the style of text may include both font recognition and language recognition. In one example, language recognition may be performed on an input text image based on a trained neural network capable of recognizing the language, and language information associated with the input text image may be output; then, combining the language information, carrying out font identification on the input character image based on the trained neural network capable of identifying the language font, and outputting font information associated with the input character image; this font information is then used for subsequent text recognition. For example, the method for training the neural network for recognizing languages is substantially the same as the method for training the neural network for recognizing fonts in the foregoing embodiment, and is not repeated herein for brevity. In another example, the font identification may be performed first, then the language identification may be performed by combining the font information, and finally the language information may be combined for the subsequent character identification. In other examples, font identification and language identification may be performed without being separated, and then subsequent character identification may be performed in combination with the font information and the language information.
Although the above examples are described in terms of recognition of fonts and languages of text, they are merely exemplary and may also include recognition of other text styles. Furthermore, it should be understood that the present invention is not limited to the specific character style recognition method, and that the present invention can be applied to the character recognition method according to the embodiment of the present invention, regardless of the existing character style recognition method or the character style recognition method developed in the future, and the present invention is also included in the protection scope of the present invention.
In step S220, a character recognition database corresponding to the character style information is selected from the trained plurality of character recognition databases for different character styles for performing character recognition on the character image.
In one embodiment, different word recognition databases may be trained based on different word styles. For example, in one example, a text recognition database corresponding to different fonts, respectively, may be trained. In another example, a character recognition database corresponding to different languages, respectively, may be trained. In other examples, a character recognition database corresponding to other different character styles, respectively, may be trained.
Based on the character style information associated with the input character image obtained in step S210, a character recognition database corresponding to the character style information may be selected from the trained character recognition databases for character recognition. Because the character recognition database used for character recognition is specially used for inputting the character style of the character image, the recognition efficiency is improved, and the recognition accuracy is improved.
In one embodiment, the text style information output in step S210 may include information of a plurality of selectable text styles similar to the text style in the input text image, that is, the output text style information includes information of more than one text styles, which are similar to the text styles in the input text image or have the similarity degree ranked in the first few digits. Based on these alternative text style information, certain processing may be performed to select the most similar text style information for subsequent text recognition. For example, the selectable character style with the maximum corresponding similarity is selected as the recognition result to be output, or the selectable character style with the corresponding similarity larger than a preset threshold is selected as the recognition result to be output.
For example, in step S220, a character recognition database corresponding to information of a character style with the highest similarity of character styles in the input character image may be selected for character recognition based on the selectable character style information. Alternatively, the character recognition database corresponding to each of the selectable character style information may be selected for performing multiple character recognition, the corresponding selectable character recognition result may be output, and then the selectable character recognition results may be processed to obtain a final character recognition result.
Regardless of the method, because the optimal database can be automatically selected according to the character style information instead of the single character recognition database, confusion among similar fonts with different character styles can be effectively avoided, and the recognition rate is improved because the similar fonts with different character styles are prevented from being distinguished.
Based on the above description, according to the character recognition method based on character style recognition of the embodiment of the present invention, before character recognition, character style recognition is performed, and the character recognition database of the character style is selected for character recognition based on different character styles, so that not only the recognition efficiency can be improved, but also the recognition accuracy can be improved.
For example, a method for word recognition based on word style recognition according to an embodiment of the present invention may be implemented in a device, apparatus or system having a memory and a processor.
The character recognition method based on character style recognition according to the embodiment of the invention can be deployed at personal terminals such as smart phones, tablet computers, personal computers and the like. Alternatively, the character recognition method based on character style recognition according to the embodiment of the present invention may also be deployed at a server (or cloud). Alternatively, the character recognition method based on character style recognition according to the embodiment of the present invention may also be distributively deployed at the server side (or cloud side) and the personal terminal side.
Fig. 3 shows a schematic block diagram of a text recognition apparatus 300 based on text style recognition according to an embodiment of the present invention.
As shown in fig. 3, a character recognition apparatus 300 based on character style recognition according to an embodiment of the present invention includes a character style recognition module 310 and a character recognition module 320.
The character style recognition module 310 is configured to recognize a character style of an input character image and output character style information associated with the character image. The character recognition module 320 is configured to select a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for performing character recognition on the character image. Both the text style recognition module 310 and the text recognition module 320 may be implemented by the processor 102 in the electronic device shown in fig. 1 executing program instructions stored in the storage 104.
According to the embodiment of the invention, the input character image can be a character image acquired by an image acquisition device or a character image from other sources. The text image may be a text picture, a text video, or the like. The text style of the text image may include, but is not limited to, the font of the text, the language of the text, the object on which the text is presented, various types of artistic words, and the like. The object presenting the text is, for example, paper, stone, wood, or other possible objects.
For characters with different character styles, even the same character can have different appearances, and different characters can have similar appearances due to different character styles. Taking the font as an example, the date in a relatively flat font is difficult to distinguish from the date in a common font. Therefore, the recognition rate can be effectively improved by recognizing the character style before the character recognition.
According to the embodiment of the present invention, the character style recognition module 310 may perform character style recognition on the input character image based on the trained neural network. Taking a font as an example, the text style recognition module 310 may include a font recognition module (not shown in fig. 3), which may perform font recognition on the input text image based on the trained neural network. The neural network is convenient to train, stable in performance and strong in universality, so that the character style of the input character image is recognized simply, effectively and easily on the basis of the trained neural network. In other embodiments, the font recognition module may also perform text style recognition on the input text image based on classical Optical Character Recognition (OCR).
The character style recognition module 310 may output character style information associated with the input character image through character style recognition. For example, taking a font as an example, the font identification module may output font information associated with the input text image by performing font identification on the input text image. Similarly, for example, the text style recognition module 310 may include a language recognition module (not shown in fig. 3) that can output language information associated with the input text image by performing language recognition on the input text image.
According to an embodiment of the present invention, the text style recognition module 310 may include both a font recognition module (not shown in fig. 3) and a language recognition module (not shown in fig. 3).
In one example, the language identification module may perform language identification on the input text image based on a trained neural network capable of identifying the language, and output language information associated with the input text image; then, the font identification module carries out font identification on the input character image based on the trained neural network capable of identifying the language font in combination with the language information, and outputs font information associated with the input character image; this font information is then used by the text recognition module 320 for subsequent text recognition.
In another example, the font recognition module may perform font recognition first, then the language recognition module performs language recognition by combining the font information, and finally the character recognition module 320 combines the language information for subsequent character recognition.
In other examples, the font identification module and the language identification module may perform font identification and language identification, respectively, without being sequentially performed, and then the character identification module 320 performs subsequent character identification by combining the font information and the language information.
Although the above example is described in terms of recognition by a font recognition module and a language recognition module, it is merely exemplary and the text style recognition module 310 may also include other text style recognition modules.
According to the embodiment of the invention, different character recognition databases can be trained based on different character styles. For example, in one example, a text recognition database corresponding to different fonts, respectively, may be trained. In another example, a character recognition database corresponding to different languages, respectively, may be trained. In other examples, a character recognition database corresponding to other different character styles, respectively, may be trained.
Based on the text style information associated with the input text image obtained from the text style recognition module 310, the text recognition module 320 may select a text recognition database corresponding to the text style information from the trained text recognition databases for text recognition. Because the character recognition database used for character recognition is specially used for inputting the character style of the character image, the recognition efficiency is improved, and the recognition accuracy is improved.
According to the embodiment of the present invention, the text style information output by the text style identification module 310 may include information of a plurality of selectable text styles similar to the text style in the input text image, that is, the text style information output by the text style identification module 310 includes information of more than one text styles, which are similar to the text styles in the input text image or are ranked in the top several digits. Based on these alternative text style information, a certain process may be performed to select the most similar text style information for the text recognition module 320 to perform subsequent text recognition.
Alternatively, the character recognition module 320 may select, from among the selectable character style information, a character recognition database corresponding to information of a character style with the highest similarity in character style in the input character image for character recognition. Alternatively, the character recognition module 320 may also select a character recognition database corresponding to each of the selectable character style information to perform character recognition for multiple times, output a corresponding selectable character recognition result, and perform certain processing on the selectable character recognition results to obtain a final character recognition result.
Regardless of the method, because the optimal database can be automatically selected according to the character style information instead of the single character recognition database, confusion among similar fonts with different character styles can be effectively avoided, and the recognition rate is improved because the similar fonts with different character styles are prevented from being distinguished.
Based on the above description, the character recognition device based on character style recognition according to the embodiment of the present invention performs character style recognition before performing character recognition, and selects the character recognition database of the character style for character recognition based on different character styles, so that not only the recognition efficiency but also the recognition accuracy can be improved.
Those of ordinary skill in the art will appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
FIG. 4 shows a schematic block diagram of a text-style-recognition-based text recognition system 400 in accordance with an embodiment of the present invention. The word-style recognition based word recognition system 400 includes a storage device 410 and a processor 420.
The storage device 410 stores program codes for implementing corresponding steps in the character recognition method based on character style recognition according to the embodiment of the present invention. The processor 420 is configured to run the program codes stored in the storage device 420 to perform the corresponding steps of the character recognition method based on character style recognition according to the embodiment of the present invention, and is configured to implement the corresponding modules in the character recognition device based on character style recognition according to the embodiment of the present invention. In addition, the text-style-recognition-based text recognition system 400 may further include an image capture device (not shown in fig. 4) that may be used to capture text images. Of course, the image capture device is not required and may receive input of textual images directly from other sources.
In one embodiment, the program code, when executed by processor 420, causes word style recognition based word recognition system 400 to perform the steps of: recognizing the character style of an input character image, and outputting character style information associated with the character image; and selecting a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for carrying out character recognition on the character image.
In one embodiment, the text style includes at least one of: the font of the characters, the language of the characters and the object for presenting the characters.
In one embodiment, the recognition of the character style of the input character image is based on a trained neural network.
In one embodiment, the text style information includes information of a plurality of selectable text styles similar to the text style in the text image.
In one embodiment, the selecting the character recognition database corresponding to the character style information comprises selecting the character recognition database corresponding to the information of the character style with the highest similarity of character styles in the character image.
In addition, according to an embodiment of the present invention, a storage medium is further provided, on which program instructions are stored, and when the program instructions are executed by a computer or a processor, the program instructions are used to execute corresponding steps of the character recognition method based on character style recognition according to an embodiment of the present invention, and are used to implement corresponding modules in the character recognition device based on character style recognition according to an embodiment of the present invention. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM), a portable compact disc read only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer readable storage medium can be any combination of one or more computer readable storage media, such as one containing computer readable program code for performing text style recognition on an input text image and outputting text style information associated with the text image, another containing computer readable program code for selecting a text recognition database corresponding to the text style information among a plurality of trained text recognition databases for different text styles for performing text recognition on the text image.
In one embodiment, the computer program instructions may implement the functional modules of the character recognition apparatus based on character style recognition according to the embodiment of the present invention when being executed by a computer, and/or may execute the character recognition method based on character style recognition according to the embodiment of the present invention.
In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the steps of: recognizing the character style of an input character image, and outputting character style information associated with the character image; and selecting a character recognition database corresponding to the character style information from a plurality of trained character recognition databases for different character styles for carrying out character recognition on the character image.
In one embodiment, the text style includes at least one of: the font of the characters, the language of the characters and the object for presenting the characters.
In one embodiment, the recognition of the character style of the input character image is based on a trained neural network.
In one embodiment, the text style information includes information of a plurality of selectable text styles similar to the text style in the text image.
In one embodiment, the selecting the character recognition database corresponding to the character style information comprises selecting the character recognition database corresponding to the information of the character style with the highest similarity of character styles in the character image.
The modules in the text-style-recognition-based text recognition apparatus according to the embodiment of the present invention can be implemented by a processor of a text-style-recognition-based electronic device according to the embodiment of the present invention executing computer program instructions stored in a memory, or can be implemented when computer instructions stored in a computer-readable storage medium of a computer program product according to the embodiment of the present invention are executed by a computer.
According to the character recognition method, the device and the system based on character style recognition and the storage medium provided by the embodiment of the invention, the character style is recognized before character recognition is carried out, and the character recognition database of the character style is selected for character recognition based on different character styles, so that the recognition efficiency can be improved, and the recognition accuracy can be improved.
Although the illustrative embodiments have been described herein with reference to the accompanying drawings, it is to be understood that the foregoing illustrative embodiments are merely exemplary and are not intended to limit the scope of the invention thereto. Various changes and modifications may be effected therein by one of ordinary skill in the pertinent art without departing from the scope or spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as set forth in the appended claims.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the units is only one logical functional division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another device, or some features may be omitted, or not executed.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. However, the method of the present invention should not be construed to reflect the intent: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
It will be understood by those skilled in the art that all of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where such features are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It will be appreciated by those skilled in the art that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some of the modules in an item analysis apparatus according to embodiments of the present invention. The present invention may also be embodied as apparatus programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
The above description is only for the specific embodiment of the present invention or the description thereof, and the protection scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. A character recognition method based on character style recognition is characterized by comprising the following steps:
recognizing the character style of an input character image, and outputting character style information associated with the character image; and
training different character recognition databases based on different character styles, selecting a character recognition database corresponding to the character style information from a plurality of character recognition databases trained for different character styles for performing character recognition on the character image,
the character styles comprise character fonts, character languages and character presenting objects, the method further comprises the steps of recognizing one character style, and then recognizing the other character style by combining recognition results of the one character style, wherein the recognition of the character style of the input character image is based on a trained neural network.
2. The character recognition method of claim 1, wherein the character style information includes information of a plurality of selectable character styles similar to a character style in the character image.
3. The method of claim 2, wherein selecting the text recognition database corresponding to the text style information comprises selecting the text recognition database corresponding to the information of the text style with the highest similarity of text styles in the text image.
4. A character recognition apparatus based on character style recognition, the character recognition apparatus comprising:
the character style recognition module is used for recognizing the character style of the input character image and outputting character style information associated with the character image; and
a character recognition module for training different character recognition databases based on different character styles, selecting a character recognition database corresponding to the character style information from a plurality of character recognition databases trained for different character styles for character recognition of the character image,
the character recognition device is also configured to recognize one character style first and then recognize another character style by combining the recognition result of the one character style, and the character style recognition module recognizes the character style of the input character image based on a trained neural network.
5. The character recognition apparatus of claim 4, wherein the character style information includes information of a plurality of selectable character styles similar to a character style in the character image.
6. The character recognition apparatus of claim 5, wherein the character recognition module selects a character recognition database corresponding to information of a character style with the highest similarity to the character style in the character image for character recognition of the character image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610509781.6A CN106156794B (en) | 2016-07-01 | 2016-07-01 | Character recognition method and device based on character style recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610509781.6A CN106156794B (en) | 2016-07-01 | 2016-07-01 | Character recognition method and device based on character style recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106156794A CN106156794A (en) | 2016-11-23 |
| CN106156794B true CN106156794B (en) | 2020-12-25 |
Family
ID=57350857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610509781.6A Active CN106156794B (en) | 2016-07-01 | 2016-07-01 | Character recognition method and device based on character style recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106156794B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109165376B (en) * | 2018-06-28 | 2023-07-18 | 西交利物浦大学 | Style character generation method based on small amount of samples |
| CN109308475A (en) * | 2018-07-26 | 2019-02-05 | 北京百悟科技有限公司 | A kind of character recognition method and device |
| CN109784330B (en) * | 2019-01-03 | 2021-03-23 | 北京百度网讯科技有限公司 | Signboard content identification method, device and equipment |
| CN110287951B (en) * | 2019-06-21 | 2022-04-12 | 北京百度网讯科技有限公司 | Character recognition method and device |
| CN113536771B (en) * | 2021-09-17 | 2021-12-24 | 深圳前海环融联易信息科技服务有限公司 | Element information extraction method, device, equipment and medium based on text recognition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1808468A (en) * | 2005-01-17 | 2006-07-26 | 佳能信息技术（北京）有限公司 | Optical character recognition method and system |
| CN101097600A (en) * | 2006-06-29 | 2008-01-02 | 北大方正集团有限公司 | Character recognizing method and system |
| CN102262619A (en) * | 2010-05-31 | 2011-11-30 | 汉王科技股份有限公司 | Method and device for extracting characters of document |
| CN102262731A (en) * | 2011-07-11 | 2011-11-30 | 哈尔滨工业大学 | Character recognizing method based on sparse coding |
| CN103455806A (en) * | 2012-05-31 | 2013-12-18 | 富士通株式会社 | Document processing device, document processing method and scanner |
| CN105279152A (en) * | 2014-06-24 | 2016-01-27 | 腾讯科技（深圳）有限公司 | Method and device for word-fetching translation |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8391602B2 (en) * | 2010-04-08 | 2013-03-05 | University Of Calcutta | Character recognition |
| US9152871B2 (en) * | 2013-09-02 | 2015-10-06 | Qualcomm Incorporated | Multiple hypothesis testing for word detection |
| CN104834890B (en) * | 2015-02-13 | 2018-01-05 | 浙江大学 | A kind of extracting method to word expression information in calligraphy work |
| CN105117740B (en) * | 2015-08-21 | 2021-06-15 | 北京旷视科技有限公司 | Font identification method and apparatus |
-
2016
- 2016-07-01 CN CN201610509781.6A patent/CN106156794B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1808468A (en) * | 2005-01-17 | 2006-07-26 | 佳能信息技术（北京）有限公司 | Optical character recognition method and system |
| CN101097600A (en) * | 2006-06-29 | 2008-01-02 | 北大方正集团有限公司 | Character recognizing method and system |
| CN102262619A (en) * | 2010-05-31 | 2011-11-30 | 汉王科技股份有限公司 | Method and device for extracting characters of document |
| CN102262731A (en) * | 2011-07-11 | 2011-11-30 | 哈尔滨工业大学 | Character recognizing method based on sparse coding |
| CN103455806A (en) * | 2012-05-31 | 2013-12-18 | 富士通株式会社 | Document processing device, document processing method and scanner |
| CN105279152A (en) * | 2014-06-24 | 2016-01-27 | 腾讯科技（深圳）有限公司 | Method and device for word-fetching translation |
Non-Patent Citations (1)
| Title |
|---|
| 光学字符识别;通信学会cic;《百度百科https://baike.baidu.com/history/%E5%85%89%E5%AD%A6%E5%AD%97%E7%AC%A6%E8%AF%86%E5%88%AB/72192072》;20150106;第2页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106156794A (en) | 2016-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106446816B (en) | Face recognition method and device | |
| CN106156794B (en) | Character recognition method and device based on character style recognition | |
| US10032072B1 (en) | Text recognition and localization with deep learning | |
| CN101542531B (en) | Image recognizing apparatus and image recognizing method | |
| CN106650662B (en) | Target object shielding detection method and device | |
| US20190377956A1 (en) | Method and apparatus for processing video | |
| CN109145766B (en) | Model training method and device, recognition method, electronic device and storage medium | |
| US8649602B2 (en) | Systems and methods for tagging photos | |
| CN110555372A (en) | Data entry method, device, equipment and storage medium | |
| US9239961B1 (en) | Text recognition near an edge | |
| CN106203425B (en) | Character recognition method and device | |
| WO2016015621A1 (en) | Human face picture name recognition method and system | |
| CN106339479A (en) | Picture naming method and terminal | |
| CN110111902B (en) | Acute infectious disease attack period prediction method, device and storage medium | |
| KR102373884B1 (en) | Image data processing method for searching images by text | |
| CN111709414A (en) | AR device, character recognition method and device thereof, and computer-readable storage medium | |
| CN104750791A (en) | Image retrieval method and device | |
| CN111640193A (en) | Word processing method, word processing device, computer equipment and storage medium | |
| CN111061887A (en) | News character photo extraction method, device, equipment and storage medium | |
| CN106682187B (en) | Method and device for establishing image base | |
| CN106250755B (en) | Method and device for generating verification code | |
| CN107992872B (en) | Method for carrying out text recognition on picture and mobile terminal | |
| CN110874570A (en) | Face recognition method, device, equipment and computer readable storage medium | |
| CN108875496B (en) | Pedestrian representation generation and representation-based pedestrian recognition | |
| CN112241470A (en) | Video classification method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100190 Beijing, Haidian District Academy of Sciences, South Road, No. 2, block A, No. 313 Applicant after: MEGVII INC. Applicant after: Beijing maigewei Technology Co., Ltd. Address before: 100190 Beijing, Haidian District Academy of Sciences, South Road, No. 2, block A, No. 313 Applicant before: MEGVII INC. Applicant before: Beijing aperture Science and Technology Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |